We will determine how defects in adhesion proteins of epidermal basal cells inhibit both attachment to the basement membrane (BM) and adhesion- dependent cell function(s). With this goal, we identified the adhesive BM ligand laminin 5 (epiligrin) and sequenced cDNA products of the LAMA3 gene that encode the alpha3 chain of laminin 5. Laminin 5 interacts with two integrin adhesion receptors in basal cells, alpha6beta4 in hemidesmosomes (HDs) to mediate anchorage in homeostatic tissue and with integrin alpha3beta1 to control cell migration in wounds. Cross talk between these two adhesion systems balance needs for both cell anchorage in homeostasis and motility in wounds. Severe blistering of epithelium results from incomplete formation of HDs and is caused by defects in laminin 5 in patients with the gravis form of junctional epidermolysis bullosis (JEB). Defects in expression of laminin 5, as in JEB, should disrupt the adhesion-dependent epithelial functions beyond blistering of the epithelia. We wish to construct a mouse model for human JEB by targeted disruption of the LAMA3 gene using homologous recombination. To avoid embryonic or neonatal lethality we will construct a chymeric mouse from wild type mouse blastocyts and ES cells homozygous for mutations in the LAMA3 gene or by regulation of the mutations under an inducible promoter. The resulting chymeric animals will express both wild type and defective BM into adulthood permitting studies on adhesion-dependent functions on (i) in regulating proliferation and differentiation, (ii) decreased cross talk between anchorage and motility with resulting increased cell invasion, (iii) a system for examining T cell epidermotropism associated with T cell lymphoma and graft verses host disease.